I read an interesting article on CFO.com about the usual mistakes made when building financial models.
They suggest the following:
1) Poor Segregation of Data
2) Poor Documentation of Assumptions
3) Poor Documentation of Constraints
4) Difficulties in Making Changes
5) “Now It’s Here; Now It’s Not”
6) The Presentation Readiness Problem
They then suggest some ways to improve this:
—Encouraging the separation of data, and documenting the reference
sources.
—Making robust and useful key interim calculations and analyses for
which the data and your assumptions are being used.
—Encouraging a summary design for communication and presentation.
It is exactly these issues that we have spent the last few years building software to solve. We work to build all the important separations between inputs and outputs, we have over 20,000 checksums to provide integrity and then we layer a neat presentation pack on the top.
As I discussed in a previous post why is it that there is a standard for how to prepare and present historic data but no such standard in the world of forecasting? A well designed standard and training courses to supplement it, with industry body backing would surely take a lot of the pain out of the forecasting process and reduce the possibility of serious errors.
If anyone has any other ideas for how to improve the process throw them into a comment on this blog and lets discuss.
Heres a link to the CFO article http://www.cfo.com/article.cfm/11288290/c_11362095?f=singlepage
Filed under: Modelling | Tagged: excel, excel errors, forecasting, forecasting standard, modelling errors, spreadsheet inaccuracy
